← Back to community benchmarks
gemma-4-31B-it-OptiQ
Performance
16k
tokens
242.8
PP tok/s
22.4
TG tok/s
67475
TTFT (ms)
26.6
Peak mem (GB)
Hardware
Chip
M3 Ultra (60c)
Memory
256 GB
GPU Cores
60
Software
oMLX
v0.3.8
macOS
macOS 26.5
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 251.3 | 25.7 | 22.3 GB | view |
| 4k | 253.9 | 24.8 | 24.9 GB | view |
| 8k | 251.3 | 24.1 | 25.5 GB | view |
| 16k | 242.8 | 22.4 | 26.6 GB | current |
| 32k | 226.2 | 19.5 | 29.6 GB | view |
| 64k | 197.7 | 15.4 | 35.5 GB | view |
| 128k | 151.2 | 9.0 | 47.6 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 25.7 | 1.00× |
| 2× | 39.8 | 1.55× |
| 4× | 47.3 | 1.84× |